An Empirical Study on Hyperparameter Optimization for Fine-Tuning Pre-trained Language Models

• URL: http://arxiv.org/abs/2106.09204v1
• Date: Thu, 17 Jun 2021 01:58:32 GMT
• Title: An Empirical Study on Hyperparameter Optimization for Fine-Tuning Pre-trained Language Models
• Abstract summary: We study and report three HPO algorithms' performances on fine-tuning two state-of-the-art language models on the GLUE dataset. We find that using the same time budget, HPO often fails to outperform grid search due to two reasons: insufficient time budget and overfitting.
• Score: 10.599060942412144
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The performance of fine-tuning pre-trained language models largely depends on the hyperparameter configuration. In this paper, we investigate the performance of modern hyperparameter optimization methods (HPO) on fine-tuning pre-trained language models. First, we study and report three HPO algorithms' performances on fine-tuning two state-of-the-art language models on the GLUE dataset. We find that using the same time budget, HPO often fails to outperform grid search due to two reasons: insufficient time budget and overfitting. We propose two general strategies and an experimental procedure to systematically troubleshoot HPO's failure cases. By applying the procedure, we observe that HPO can succeed with more appropriate settings in the search space and time budget; however, in certain cases overfitting remains. Finally, we make suggestions for future work. Our implementation can be found in https://github.com/microsoft/FLAML/tree/main/flaml/nlp/.

Related papers

• POCAII: Parameter Optimization with Conscious Allocation using Iterative Intelligence [4.478575931884855]
  POCAII is a flexible scheme for managing a hyper parameter optimization budget.<n>It shows superior performance in low-budget hyper parameter optimization regimes.<n>It has wide applications to real-world problems.
  arXiv  Detail & Related papers  (2025-05-16T23:05:07Z)
• ULTHO: Ultra-Lightweight yet Efficient Hyperparameter Optimization in Deep Reinforcement Learning [50.53705050673944]
  We propose ULTHO, an ultra-lightweight yet powerful framework for fast HPO in deep RL within single runs.<n>Specifically, we formulate the HPO process as a multi-armed bandit with clustered arms (MABC) and link it directly to long-term return optimization.<n>We test ULTHO on benchmarks including ALE, Procgen, MiniGrid, and PyBullet.
  arXiv  Detail & Related papers  (2025-03-08T07:03:43Z)
• HyperQ-Opt: Q-learning for Hyperparameter Optimization [0.0]
  This paper presents a novel perspective on HPO by formulating it as a sequential decision-making problem and leveraging Q-learning, a reinforcement learning technique. The approaches are evaluated for their ability to find optimal or near-optimal configurations within a limited number of trials. By shifting the paradigm toward policy-based optimization, this work contributes to advancing HPO methods for scalable and efficient machine learning applications.
  arXiv  Detail & Related papers  (2024-12-23T18:22:34Z)
• Hierarchical Preference Optimization: Learning to achieve goals via feasible subgoals prediction [71.81851971324187]
  This work introduces Hierarchical Preference Optimization (HPO), a novel approach to hierarchical reinforcement learning (HRL) HPO addresses non-stationarity and infeasible subgoal generation issues when solving complex robotic control tasks. Experiments on challenging robotic navigation and manipulation tasks demonstrate impressive performance of HPO, where it shows an improvement of up to 35% over the baselines.
  arXiv  Detail & Related papers  (2024-11-01T04:58:40Z)
• Improving Hyperparameter Optimization with Checkpointed Model Weights [16.509585437768063]
  In this work, we propose an HPO method for neural networks using logged checkpoints of the trained weights. Our method, Forecasting Model Search (FMS), embeds weights into a Gaussian process deep kernel surrogate model.
  arXiv  Detail & Related papers  (2024-06-26T17:59:54Z)
• From $r$ to $Q^*$: Your Language Model is Secretly a Q-Function [50.812404038684505]
  We show that we can derive DPO in the token-level MDP as a general inverse Q-learning algorithm, which satisfies the Bellman equation. We discuss applications of our work, including information elicitation in multi-turn dialogue, reasoning, agentic applications and end-to-end training of multi-model systems.
  arXiv  Detail & Related papers  (2024-04-18T17:37:02Z)
• PriorBand: Practical Hyperparameter Optimization in the Age of Deep Learning [49.92394599459274]
  We propose PriorBand, an HPO algorithm tailored to Deep Learning (DL) pipelines. We show its robustness across a range of DL benchmarks and show its gains under informative expert input and against poor expert beliefs.
  arXiv  Detail & Related papers  (2023-06-21T16:26:14Z)
• Two-step hyperparameter optimization method: Accelerating hyperparameter search by using a fraction of a training dataset [0.15420205433587747]
  We present a two-step HPO method as a strategic solution to curbing computational demands and wait times. We present our recent application of the two-step HPO method to the development of neural network emulators for aerosol activation.
  arXiv  Detail & Related papers  (2023-02-08T02:38:26Z)
• Towards Learning Universal Hyperparameter Optimizers with Transformers [57.35920571605559]
  We introduce the OptFormer, the first text-based Transformer HPO framework that provides a universal end-to-end interface for jointly learning policy and function prediction. Our experiments demonstrate that the OptFormer can imitate at least 7 different HPO algorithms, which can be further improved via its function uncertainty estimates.
  arXiv  Detail & Related papers  (2022-05-26T12:51:32Z)
• Efficient Nearest Neighbor Language Models [114.40866461741795]
  Non-parametric neural language models (NLMs) learn predictive distributions of text utilizing an external datastore. We show how to achieve up to a 6x speed-up in inference speed while retaining comparable performance.
  arXiv  Detail & Related papers  (2021-09-09T12:32:28Z)
• Hyperparameter Optimization: Foundations, Algorithms, Best Practices and Open Challenges [5.139260825952818]
  This paper reviews important HPO methods such as grid or random search, evolutionary algorithms, Bayesian optimization, Hyperband and racing. It gives practical recommendations regarding important choices to be made when conducting HPO, including the HPO algorithms themselves, performance evaluation, how to combine HPO with ML pipelines, runtime improvements, and parallelization.
  arXiv  Detail & Related papers  (2021-07-13T04:55:47Z)
• Multi-objective Asynchronous Successive Halving [10.632606255280649]
  We propose algorithms that extend successive asynchronous halving (ASHA) to the multi-objective (MO) setting. Our empirical analysis shows that MO ASHA enables to perform MO HPO at scale. Our algorithms establish new baselines for future research in the area.
  arXiv  Detail & Related papers  (2021-06-23T19:39:31Z)
• Cost-Efficient Online Hyperparameter Optimization [94.60924644778558]
  We propose an online HPO algorithm that reaches human expert-level performance within a single run of the experiment. Our proposed online HPO algorithm reaches human expert-level performance within a single run of the experiment, while incurring only modest computational overhead compared to regular training.
  arXiv  Detail & Related papers  (2021-01-17T04:55:30Z)
• Practical and sample efficient zero-shot HPO [8.41866793161234]
  We provide an overview of available approaches and introduce two novel techniques to handle the problem. The first is based on a surrogate model and adaptively chooses pairs of dataset, configuration to query. The second is for settings where finding, tuning and testing a surrogate model is problematic, is a multi-fidelity technique combining HyperBand with submodular optimization.
  arXiv  Detail & Related papers  (2020-07-27T08:56:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.